Therapeutic Hypothermia for Comatose Cardiac Arrest Survivors

September 2007

Therapeutic Hypothermia for Comatose Cardiac Arrest Survivors

Search Strategy: Using the term “cardiac arrest” AND “hypothermia” you conduct a narrow, specific therapeutic clinical query using PUBMED with 32 “hits”. You note one clinical trial reported in the EM literature dating back to 1997 in Annals of EM as well as two large randomized controlled trials from 2002 which seem not to have yet been replicated, so you search Web of Science for subsequent literature which has cited these seminal articles. All of your evidence is identified using these two search strategies.

Working the trauma bay on a busy Monday overnight shift you receive the following page: “65y/o male HTN, “heart hx”, pulseless/vfib on arrival, shock x1 by EMS with pulse return, intubated, now with vitals 85, 110/p, to TCC3 in 5 min”

The patient actually arrives as you receive this page and EMS begins their report. Mr. B. is a 65 year old with HTN, CAD, CABG, who was at home when his wife witnessed him suddenly slumped over in his chair. Bystander CPR was not initiated. When EMS arrived 10 minutes after the event, the following rhythm was noted:

The paramedics started bagging Mr. B and then shocked him with 360J. They noted spontaneous return of pulse to the mid 80’s, strong and bounding with a palpable BP of 120. They also noted that the patient was not responsive, his pupils were fixed at 5mm, and he was not protecting his airway. EMS then intubated Mr. B. without any RSI medications. Mr. B’s neurological exam on presentation remains unchanged and his pulse continues in the mid 70’s with a BP in the 120/70 range. The tech hands you his current ECG:

After sending all of the requisite labs, you are presenting to your attending while trying to formulate your acute treatment plan. The most congenial EP fellow (who happens to be interrogating a pacemaker in a patient next door) overhears your conversation. He interjects that where he went to residency they started treating patient’s with cardiac arrest with hypothermia to optimize their neurological outcomes. You have heard about this vaguely from your crazy, obsessive-compulsive EBM attending and page him with the following PICO question:

PICO Question

Population: Unresponsive adults following cardiac arrest with ROSC (return of spontaneous circulation)

Intervention: Therapeutic cooling/Hypothermia

Comparison: No cooling (Normothermia/hyperthermia)

Outcome: Mortality, neurological morbidity, post-arrest arrhythmias, other end organ viability/damage, cost-effectiveness, bleeding complications


First years: Hypothermia for neuroprotection after cardiac arrest: Systematic review and individual patient data meta-analysis, Crit Care Med 2005; 33: 414-418. (

Second years: From evidence to clinical practice: effective implementation of therapeutic hypothermia to improve patient outcome after cardiac arrest. Crit Care Med 2006; 34: 1865-1873. (

Third years: Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. NEJM 2002; 346: 557-563. (

Fourth years: Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. NEJM 2002; 346: 549-556. (


Article 1: Hypothermia for neuroprotection after cardiac arrest: Systematic review and individual patient data meta-analysis, Crit Care Med 2005;33:414-418

Article 2: From evidence to clinical practice: Effective implementation of therapeutic hypothermia to improve patient outcome after cardiac arrest, Crit Care Med 2006;34(7):1865-1873

Article 3: Treatment of Comatose Survivors of Out-of-Hospital Cardiac Arrest with Induced Hypothermia, N Engl J Med 2002;346:557-563

Article 4: Mild Therapeutic Hypothermia to Improve the Neurologic Outcome after Cardiac Arrest, N Engl J Med 2002; 346: 549-556

Bottom Line

Three small, adequately powered controlled trials with heterogeneous patient populations, cooling methods, and random allocation schemes consistently demonstrate improved neurologically intact hospital-discharge and six-month survival in ventricular fibrillation cardiac arrest victims immediately treated with hypothermia to 32-35 degrees C for 4-24 hours with NNT = 6 and impressively narrow confidence intervals. This benefit is maintained when controlling for confounding variables including patient-specific characteristics (age, time to ROSC), method of cooling and center in which treatment occurred. Future research should assess quality of life measures and delineate the optimal method and duration of therapeutic hypothermia.

At BJH the barriers to implementing this best-evidence practice are many:

  1. lack of Awareness (beyond CCU, Critical Care, and EM physician and nursing staff, few physicians are aware of this cheap means of improving outcomes). You can do your own informal survey through the hospital asking critical care and non-critical care faculty about their knowledge of therapeutic hypothermia (evidence for, treatment efficacy, arguments against, etc.)
  2. Acceptance – among those who are aware, many refuse to accept the evidence as adequate or conclusive. This is not unique to BJH. At UCLA, the EM Abstracts guru, Jerry Hoffman, emphatically states “There is no evidence supporting Therapeutic Hypothermia” and does not advocate this practice at this point and time, despite the lack of any evidence suggesting harm or alternative therapeutic options.
  3. Applicable – some feel the extensive exclusion criteria in these trials precludes their use on all (not just most) post-arrest comatose survivors.
  4. Able – nursing feels that the hospital does not support the ED-based use of this therapy as evidenced by a) lack of our own cooling equipment; b) recent refusal of NICU to loan us their cooling equipment because somebody (not EM) misplaced it in the past. I would argue that each of the three trials we evaluated today did not use expensive cooling equipment. Instead, they used ice cubes and cooling blankets. The question I posed was what evidence supports the use of expensive cooling devices? Jonas Cooper, CCU Fellow, stated that no evidence (non-lab animal) evidence currently supports the use of the expensive cooling devices, but BJH clinical faculty are worried about slower, more erratic cooling rates with ice cubes and cooling blankets.
  5. Act Upon – even if one is aware of and accepts the evidence, finds it applicable to their patient population, and is able to replicate the study protocol, we still have to remember to do so when the situation arises. Some of the “act upon” barriers are memory recall, but others are support from the institution to implement this protocol when the ED is overcrowded and multiple high-acuity, time-consuming patients present simultaneously. If the CCU and Neurology believe this therapeutic resuscitation option is beneficial and worthy of “standard of care” status, support systems should be in place to augment ED efforts when our personnel (physician, nursing) cannot support this time-intensive intervention.